We propose to comprehensively evaluate a series of inflammatory, antigenic, and endothelial biomarkers for type 2 diabetes mellitus among participants in the Western New York Study. We will recall nearly 3000 cohort members and identify converters to type 2 DM. Using a case-control approach, we will test the following Specific Aims: 1) determine if biomarkers of chronic inflammation including CRP, TNFalpha, IL6 predict incident DM; 2) evaluate whether markers of endothelial function including E-selectin, sICAM1, and sVCAM1 predict DM; and 3) determine if antibodies to glutamic acid decarboxylase (GAD) and IA2 identify a subtype of phenotypic type 2 diabetic persons. We hypothesize that chronic dysregulation of the immune system is a potent predictor of type 2 DM. Cases will be persons who have developed type 2 DM since their baseline examination in 1996- 1998. We propose to control for sex, race, year of baseline exam, and baseline glycemic category in the design phase of the study. Controls will be matched 3:1 to the cases on these variables. Baseline information on risk factors including age, body fat, lipoproteins, physical activity, family history of DM, and smoking will be used to evaluate confounding and effect modification. Baseline plasma samples that will provide the exposure information have been collected, handled, and frozen with a high level of standardization and precision with immediate on-site centrifugation and freezing in our Biological Specimen Bank. The exposures proposed represent state-of-the-art measures and will be assayed in laboratories that serve as CORE labs for several national studies. The proposed research is of great importance due to the epidemic nature of type 2 DM, and the promise that it may be preventable. A better understanding of the etiology of type 2 and a better definition of the "at risk" phenotype can provide health benefits for the millions at risk for acquiring DM.